Question

In a case-control study of a suspected association between breast cancer and the contraceptive pill, all of the following are true statements EXCEPT:

• A. The control should come from a population that has the same potential for breast cancer as the cases.
• B. The control should exclude women known to be taking the pill at the time of the survey.
• C. All the controls need to be healthy.
• D. The attributable risk of breast cancer resulting from the pill may be directly measured.

Hint:

Case control studies fix the number of cases and controls, so they give odds ratio, not incidence or attributable risk.

Answer 1

The Correct Option is D

This question tests the basic design rules of a case control study and what it can and cannot measure. Let's go through each option.

1. The control should come from a population that has the same potential for breast cancer as the cases: correct. If the control group and case group don't share the same background risk profile, the comparison of pill use gets biased.
2. The control should exclude women known to be taking the pill at the time of the survey: correct. Keeping current pill users out of the control group avoids confusion between past and present exposure and keeps the exposure history comparison fair.
3. All the controls need to be healthy: correct. Controls are chosen precisely because they do not have breast cancer, so this is expected of any control group in this design.
4. The attributable risk of breast cancer resulting from the pill may be directly measured: incorrect. A case control study starts by fixing the number of cases and controls, so it never captures the true incidence of breast cancer in pill users versus non users. Without incidence figures, attributable risk (the extra disease burden caused by the exposure) cannot be worked out directly. Only the odds ratio, a stand in for relative risk, is available from this design.

The statement that does not hold is the last one, since a case control study cannot directly give attributable risk. It can only estimate relative risk through the odds ratio.

Let's summarize:

• Case control studies compare past exposure between diseased cases and disease free controls chosen from the same source population.
• They yield an odds ratio (an estimate of relative risk), never a direct measure of incidence or attributable risk.

So the exception is: the attributable risk of breast cancer resulting from the pill may be directly measured.